SUMMARY/ABSTRACT Neuroblastoma (NB) remains a leading cause of childhood cancer deaths, and the children who do survive are left with long-term side effects, many of which can be life threatening. In this era of more precise therapies, considerable efforts are being made to identify optimal targets. While the paradigm of molecularly targeted therapies holds great promise, genomic studies have revealed that NBs are characterized by extensive intratumor genetic heterogeneity, with subclonal oncogenic drivers often selected for during standard chemoradiotherapy. Our group discovered gain-of-function mutations in the ALK receptor tyrosine kinase as the etiology for familial NB, and at the same time co-discovered with several other groups identical mutations as the most frequent somatic single nucleotide variants leading to a potent oncogenic driver in up to 15% of newly diagnosed high-risk cases. Our more recent work has shown that activating mutations in the ALK-RAS- MAPK pathway are highly enriched in the relapse NB genome, providing the impetus for deep and comprehensive characterization of the subclonal landscape of genes within these pathways across the continuum of therapy. This serves at the motivation for this Project and provides the opportunity to both adapt therapeutic approaches as tumors evolve, and also target subclonal mutations earlier to prevent the acquisition of chemotherapy resistant dominant clones. The central hypothesis to be explored here is that high-risk NBs are characterized by extensive intratumoral and stroma-derived heterogeneity and harbor pre-existing and acquired subclonal populations that confer therapy resistance that can exploited with rationally selected targeted agents. We will test our central hypothesis in three Specific Aims: 1) Define the frequency and clinical significance of subclonal driver mutations; 2) Identify therapeutic vulnerabilities imparted by inhibition of oncogenic ALK and/or RAS-MAPK signaling; 3) Target tumor cell intrinsic and extrinsic oncogenic vulnerabilities for development of rational novel therapeutics. The first Aim will employ a custom ultra-deep sequencing platform to define the clonal and subclonal architecture and mutational landscape in diagnostic and relapse NBs, including PDX models. Aim 2 is devoted to defining therapeutically exploitable oncogenic vulnerabilities with a focus on demonstrating that inhibition of FAK leads to robust anti-tumor activity in ALK- and RAS-driven NBs treated with inhibitors of these pathways. The final Aim will garner the preclinical justification required to move combination therapies to the clinic, building on our extensive preliminary data of synergistic drug interactions in our oncogene-driven models. We consider this project significant because it will result in new mechanism-based biomarker-defined therapeutic strategies that ultimately should significantly improve high-risk NB patient outcomes. This will address the major unmet need that despite unprecedented discoveries in defining the basic mechanisms of NB tumorigenesis, this knowledge has not yet translated into significantly improved outcomes for patients with high-risk disease.